## UPF3A
- **UniProt:** Q9H1J1 · **batch:** proteostasis-batch-2026-06-07c · **review status:** COMPLETE
- **PN placement:** `Translation|Cytosolic translation|Translation termination|Modulation of termination` ; **PN-node mapping:** type `no_mapping`; group `mapped, ok_for_propagation` → GO:0006415 translational termination; class/branch `context_only, too_broad`. Projected: GO:0006415 (goa_status=new_to_goa).
- **Consistency:** Gene-level consistent and notably careful on directionality: deep-research/notes and review YAML agree UPF3A is the partial NMD ANTAGONIST of UPF3B — only marginally active in canonical NMD, competes with UPF3B for the UPF2 MIF4G-III surface, sequesters UPF2 to repress/buffer NMD. The review encodes this with a NEGATED GO:0000184 (NMD, NOT, IDA PMID:16601204, ACCEPT) and an ACCEPTed GO:2000623 (negative regulation of NMD), plus core_function GO:0140311 protein sequestering activity. This antagonist directionality is the opposite of the simple "termination factor" the PN node implies — a real framing mismatch.
- **PN story / NEW pressure:** PN groups UPF3A with the active termination/NMD factors and projects GO:0006415 as new_to_goa. This both (a) over-reaches as with the other UPF genes (UPF3A is not a peptide-release factor), and (b) ignores UPF3A's antagonist/negative-regulation directionality already captured by GO:2000623. No NEW term needed; the negative-regulation and sequestering story is already annotated.
- **Mapping strategy:** GROUP→GO:0006415 is wrong for UPF3A on two counts: wrong process (peptide release) and wrong direction (UPF3A dampens, not drives, decay). Projecting a positive translation-termination/NMD-adjacent term onto a partial antagonist is a directionality error the reviewers explicitly guarded against (negated GO:0000184). Type-level `no_mapping` should govern.
- **Evidence alignment:** PN row carries no reference titles; review cites PMID:16601204 (UPF3A inactive in canonical NMD), PMID:35640974 (UPF2-dependent stabilization), PMID:14636577 (UPF2/EJC interactions). No competing PN citations.
- **Verdict:** Consistent gene biology with correct antagonist directionality in the YAML; PN group→GO:0006415 over-reaches AND mis-directs (UPF3A is an NMD repressor — GO:2000623 — not a termination/decay driver).

**Recommended edits:** [MAP] do NOT project GO:0006415 (translational termination) onto UPF3A; UPF3A is a partial NMD antagonist (negated GO:0000184; GO:2000623 negative regulation of NMD; GO:0140311 sequestering). The type-level `no_mapping` should win, and any node-level frame must respect its negative/antagonist directionality.